Bayesian networks are directed acyclic graphs that model probability distributions over a set of random variables.

Each node in the network represents a variable, and the edges indicate conditional dependencies between these variables.

The probability distribution is represented by the product rule of conditional probabilities, based on the parent nodes of each node.

Bayesian networks enable efficient inference, that is, calculating probabilities for unknown variables when some variables are observed.

They are used in many fields, such as diagnosis, prognosis, or decision-making under uncertainty.